From: Yongji Xie <xyjxie@linux.vnet.ibm.com>
To: Alexey Kardashevskiy <aik@ozlabs.ru>,
kvm@vger.kernel.org, linux-kernel@vger.kernel.org,
linux-pci@vger.kernel.org, linuxppc-dev@lists.ozlabs.org,
linux-doc@vger.kernel.org
Cc: bhelgaas@google.com, corbet@lwn.net, alex.williamson@redhat.com,
benh@kernel.crashing.org, paulus@samba.org, mpe@ellerman.id.au,
warrier@linux.vnet.ibm.com, zhong@linux.vnet.ibm.com,
nikunj@linux.vnet.ibm.com
Subject: Re: [RFC PATCH v4 1/7] PCI: Add a new option for resource_alignment to reassign alignment
Date: Thu, 10 Mar 2016 12:47:41 +0800 [thread overview]
Message-ID: <56E0FC6D.6060709@linux.vnet.ibm.com> (raw)
In-Reply-To: <56E0D9BB.5010300@ozlabs.ru>
On 2016/3/10 10:19, Alexey Kardashevskiy wrote:
> On 03/07/2016 06:48 PM, Yongji Xie wrote:
>> When using resource_alignment kernel parameter, the current
>> implement reassigns the alignment by changing resources' size
>> which can potentially break some drivers.
>
> How can this possibly break any driver?... It rounds up, not down,
> what do I miss here?
>
If the driver uses the size to locate some register, e.g. the length of
register is related to the size, or get some other information of the
device. The wrong size may break this driver.
So I think it's better not to touch the size here. The resource_size()
should be the real size.
>>
>> So this patch adds a new option "noresize" for the parameter
>> to solve this problem.
>>
>> Signed-off-by: Yongji Xie <xyjxie@linux.vnet.ibm.com>
>> ---
>> Documentation/kernel-parameters.txt | 5 ++++-
>> drivers/pci/pci.c | 36
>> +++++++++++++++++++++++++----------
>> 2 files changed, 30 insertions(+), 11 deletions(-)
>>
>> diff --git a/Documentation/kernel-parameters.txt
>> b/Documentation/kernel-parameters.txt
>> index 9a53c92..d8b29ab 100644
>> --- a/Documentation/kernel-parameters.txt
>> +++ b/Documentation/kernel-parameters.txt
>> @@ -2912,13 +2912,16 @@ bytes respectively. Such letter suffixes can
>> also be entirely omitted.
>> window. The default value is 64 megabytes.
>> resource_alignment=
>> Format:
>> - [<order of align>@][<domain>:]<bus>:<slot>.<func>[;
>> ...]
>> + [<order of align>@][<domain>:]<bus>:<slot>.<func>
>> + [:noresize][; ...]
>> Specifies alignment and device to reassign
>> aligned memory resources.
>> If <order of align> is not specified,
>> PAGE_SIZE is used as alignment.
>> PCI-PCI bridge can be specified, if resource
>> windows need to be expanded.
>> + noresize: Don't change the resources' sizes when
>> + reassigning alignment.
>> ecrc= Enable/disable PCIe ECRC (transaction layer
>> end-to-end CRC checking).
>> bios: Use BIOS/firmware settings. This is the
>> diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
>> index 602eb42..760cce5 100644
>> --- a/drivers/pci/pci.c
>> +++ b/drivers/pci/pci.c
>> @@ -4598,7 +4598,8 @@ static DEFINE_SPINLOCK(resource_alignment_lock);
>> * RETURNS: Resource alignment if it is specified.
>> * Zero if it is not specified.
>> */
>> -static resource_size_t pci_specified_resource_alignment(struct
>> pci_dev *dev)
>> +static resource_size_t pci_specified_resource_alignment(struct
>> pci_dev *dev,
>> + bool *resize)
>> {
>> int seg, bus, slot, func, align_order, count;
>> resource_size_t align = 0;
>> @@ -4626,6 +4627,11 @@ static resource_size_t
>> pci_specified_resource_alignment(struct pci_dev *dev)
>> }
>> }
>> p += count;
>> + if (!strncmp(p, ":noresize", 9)) {
>> + *resize = false;
>> + p += 9;
>> + } else
>> + *resize = true;
>> if (seg == pci_domain_nr(dev->bus) &&
>> bus == dev->bus->number &&
>> slot == PCI_SLOT(dev->devfn) &&
>> @@ -4658,11 +4664,12 @@ void
>> pci_reassigndev_resource_alignment(struct pci_dev *dev)
>> {
>> int i;
>> struct resource *r;
>> + bool resize;
>> resource_size_t align, size;
>> u16 command;
>>
>> /* check if specified PCI is target device to reassign */
>> - align = pci_specified_resource_alignment(dev);
>> + align = pci_specified_resource_alignment(dev, &resize);
>
> A compiler should have warned here about passing a pointer to
> unitialized @resize.
>
OK. I'll fix it.
Thanks,
Yongji Xie
>> if (!align)
>> return;
>>
>> @@ -4684,15 +4691,24 @@ void
>> pci_reassigndev_resource_alignment(struct pci_dev *dev)
>> if (!(r->flags & IORESOURCE_MEM))
>> continue;
>> size = resource_size(r);
>> - if (size < align) {
>> - size = align;
>> - dev_info(&dev->dev,
>> - "Rounding up size of resource #%d to %#llx.\n",
>> - i, (unsigned long long)size);
>> + if (resize) {
>> + if (size < align) {
>> + size = align;
>> + dev_info(&dev->dev,
>> + "Rounding up size of resource #%d to %#llx.\n",
>> + i, (unsigned long long)size);
>> + }
>> + r->flags |= IORESOURCE_UNSET;
>> + r->end = size - 1;
>> + r->start = 0;
>> + } else {
>> + if (size > align)
>> + align = size;
>> + r->flags &= ~IORESOURCE_SIZEALIGN;
>> + r->flags |= IORESOURCE_STARTALIGN | IORESOURCE_UNSET;
>> + r->start = align;
>> + r->end = r->start + size - 1;
>> }
>> - r->flags |= IORESOURCE_UNSET;
>> - r->end = size - 1;
>> - r->start = 0;
>> }
>> /* Need to disable bridge's resource window,
>> * to enable the kernel to reassign new resource
>>
>
>
next prev parent reply other threads:[~2016-03-10 4:47 UTC|newest]
Thread overview: 27+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-03-07 7:48 [RFC PATCH v4 0/7] vfio-pci: Allow to mmap sub-page MMIO BARs and MSI-X table Yongji Xie
2016-03-07 7:48 ` [RFC PATCH v4 1/7] PCI: Add a new option for resource_alignment to reassign alignment Yongji Xie
2016-03-10 2:19 ` Alexey Kardashevskiy
2016-03-10 4:47 ` Yongji Xie [this message]
2016-03-07 7:48 ` [RFC PATCH v4 2/7] PCI: Use IORESOURCE_WINDOW to identify bridge resources Yongji Xie
2016-03-07 7:48 ` [RFC PATCH v4 3/7] PCI: Ignore resource_alignment if PCI_PROBE_ONLY was set Yongji Xie
2016-03-16 16:31 ` Alex Williamson
2016-03-17 11:35 ` Yongji Xie
2016-03-07 7:48 ` [RFC PATCH v4 4/7] PCI: Modify resource_alignment to support multiple devices Yongji Xie
2016-03-16 16:30 ` Alex Williamson
2016-03-17 11:28 ` Yongji Xie
2016-03-17 12:40 ` Alex Williamson
2016-03-18 15:04 ` Yongji Xie
2016-03-07 7:48 ` [RFC PATCH v4 5/7] vfio-pci: Allow to mmap sub-page MMIO BARs if the mmio page is exclusive Yongji Xie
2016-03-16 16:30 ` Alex Williamson
2016-03-17 11:29 ` Yongji Xie
2016-03-07 7:48 ` [RFC PATCH v4 6/7] vfio-pci: Allow to mmap MSI-X table if IOMMU_CAP_INTR_REMAP was set Yongji Xie
2016-03-16 16:31 ` Alex Williamson
2016-03-17 11:32 ` Yongji Xie
2016-03-07 7:48 ` [RFC PATCH v4 7/7] powerpc/powernv/pci-ioda: Add IOMMU_CAP_INTR_REMAP for IODA host bridge Yongji Xie
2016-03-16 16:32 ` Alex Williamson
2016-03-17 11:38 ` Yongji Xie
2016-03-17 12:48 ` Alex Williamson
2016-03-18 11:51 ` Yongji Xie
2016-03-16 10:51 ` [RFC PATCH v4 0/7] vfio-pci: Allow to mmap sub-page MMIO BARs and MSI-X table Yongji Xie
2016-03-16 14:10 ` Bjorn Helgaas
2016-03-17 10:46 ` Yongji Xie
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=56E0FC6D.6060709@linux.vnet.ibm.com \
--to=xyjxie@linux.vnet.ibm.com \
--cc=aik@ozlabs.ru \
--cc=alex.williamson@redhat.com \
--cc=benh@kernel.crashing.org \
--cc=bhelgaas@google.com \
--cc=corbet@lwn.net \
--cc=kvm@vger.kernel.org \
--cc=linux-doc@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-pci@vger.kernel.org \
--cc=linuxppc-dev@lists.ozlabs.org \
--cc=mpe@ellerman.id.au \
--cc=nikunj@linux.vnet.ibm.com \
--cc=paulus@samba.org \
--cc=warrier@linux.vnet.ibm.com \
--cc=zhong@linux.vnet.ibm.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).